Unforced Centennial Variations Simulated by Coupled GCMs

Today, blogger slimething left me a link to the Karnauskas et al (2012) paper A Pacific Centennial Oscillation Predicted by Coupled GCMs. Thanks, slimething.

I’m not going to offer any thoughts, because I don’t want to influence your comments, but I did highlight the two concluding sentences in the following abstract for you, and I’ve also reproduced the two implications listed in the Summary and concluding remarks.


Internal climate variability at the centennial time scale is investigated using long control integrations from three state-of-the-art global coupled general circulation models. In the absence of external forcing, all three models produce centennial variability in the mean zonal sea surface temperature (SST) and sea level pressure (SLP) gradients in the equatorial Pacific with counterparts in the extratropics. The centennial pattern in the tropical Pacific is dissimilar to that of the interannual El Niño–Southern Oscillation (ENSO), in that the most prominent expression in temperature is found beneath the surface of the western Pacific warm pool. Some global repercussions nevertheless are analogous, such as a hemispherically symmetric atmospheric wave pattern of alternating highs and lows. Centennial variability in western equatorial Pacific SST is a result of the strong asymmetry of interannual ocean heat content anomalies, while the eastern equatorial Pacific exhibits a lagged, Bjerknes-like response to temperature and convection in the west. The extratropical counterpart is shown to be a flux-driven response to the hemispherically symmetric circulation anomalies emanating from the tropical Pacific.

Significant centennial-length trends in the zonal SST and SLP gradients rivaling those estimated from observations and model simulations forced with increasing CO2 appear to be inherent features of the internal climate dynamics simulated by all three models. Unforced variability and trends on the centennial time scale therefore need to be addressed in estimated uncertainties, beyond more traditional signal-to-noise estimates that do not account for natural variability on the centennial time scale.

From the Summary and concluding remarks:

1) If nature exhibits such strong natural variability of tropical Pacific SSTs on centennial time scales, then assumptions that the observed trend over the past century to a century and a half is a response to radiative forcing are tenuous. It could in fact be that the observed trend over the past century and a half is merely reflective of internal variability. If so, it could strengthen or weaken in the future as the natural variability evolves. This will combine with, and potentially interact with, any forced response and thus have implications for tropical Pacific and global climate.

2) If the centennial variability in the models is spurious, then it nevertheless is a robust component of the three analyzed models, is likely to exist in other models, and therefore will continue to influence coupled GCM projections of future climate, as well as initialized decadal hindcasts and forecasts conducted with GCMs. In all cases, it must be known at what stage the natural centennial variability exists at the beginning of a forecast or projection to isolate the forced change from the modeled internal variability.


About Bob Tisdale

Research interest: the long-term aftereffects of El Niño and La Nina events on global sea surface temperature and ocean heat content. Author of the ebook Who Turned on the Heat? and regular contributor at WattsUpWithThat.
This entry was posted in Climate Model Problems. Bookmark the permalink.

13 Responses to Unforced Centennial Variations Simulated by Coupled GCMs

  1. DB says:

    Wow! Very interesting, Bob.

  2. Pascvaks says:

    I occassionally say, and often think, that we “know” nothing, but we are constantly in the process of “knowing”. I know, it’s not entirely correct; I guess I’m just more and more amazed at how much we don’t know (and how much we’ve forgotten –we don’t retain knowledge very well either;-). The biggest problem with GCM’s is all the data and all the knowledge (because we’re constatly “knowing” and just don’t “know” what we’re missing yet) that they do not contain, for some this comes as a suprise, but when you stop to think how far out we can predict the weather today in 2012, I’m suprised they are suprised. “Climatology” is the NEW Alchemy for some and they just love to shout out “Eureka!” once or twice a week, for the real scientists it’s a lifelong, often disappointing slog, with an all too rare glimmer of ‘something’ real at the end of a long dark tunnel.

  3. Bob Tisdale says:

    Pascvaks: The more detail they try to add to the climate models, the more uncertainty they realize. And they still many of the basics are unknown.

  4. Lars P. says:

    Bob Tisdale says:
    October 4, 2012 at 5:24 pm
    Pascvaks: The more detail they try to add to the climate models, the more uncertainty they realize. And they still many of the basics are unknown.
    Very interesting Bob thank you. Actually the better the data we have, the more precise the measurement, the less room there is for adjustments, the more the CAGW theory is shown to be wrong.
    The model results are often shown without uncertainty range, and even when not taking into account natural climate variations and decadal/centenial cycles. Temperature reconstructions are shown with an irrational narrow uncertainty range.
    Btw, this result will be used by warmista to explain the missing warming now, but will not be accepted to explain the past warming.
    And I bet models still do not manage to simulate LIA and MWP, even with the data where some try hard to adjust those away…

  5. Pascvaks says:

    Bob & Lars P. – Thanks, agree with you both. Was trying to say something like — ‘we have heaps of data but we’re not smart enough (yet) to know what it all means, the results of our efforts so far are more and more, and bigger and bigger data dumps, data shuffles, and gee-wiz software crunches of data, data, data, we haven’t yet made the discoveries that will tell us what all the data really means or how to use it to telling us what we want to know or if the data we have means anything at all’ — we’re long on data but short on the necessary science know-how of how to use it — or something like that?;-) Thanks again.

  6. tallbloke says:

    Given that the climatologits rely n the Sun to explain the larger centennial variations in climate prior to the human emission of co2, it’s a fair bet that the Sun is responsible for centennial variation in the C20th too. specially considering the prolonged solar maximum in the mid C20th.

    The solar variability is linked to cloud albedo variability by Nir Shaviv’s study on using the oceans as a calorimeter. Moreover, the long term millennial scale variability in solar proxies roughly matches other temperature proxies, so cloud and solar variability are proportional and roughly in phase.

    Given that reconstructions of the AMO and PDO don’t include much solar-cloud forcing, it seems likely to me that the AMO reconstruction overestimates the amplitude of the AMO around the late 1800s. This would explain the divergence of the SST record from my model, which integrates sunspot records departing from the ocean equilibrium value, the AMO, and SOI. Lower the amplitude of the positive AMO between 1880 and 1920 by 0.2C and it all fits perfectly.

    The model is generated from this combination of curves, which are amplitude scaled to Shaviv’s empirical findings for SSN, and the shorter term SST departures for AMO and SOI-ENSO.

    It’s worth noting also that ENSO is often in antiphase to the AMO in the C20th. This also supports a reduction of the AMO amplitude at the end of the 1800s.

    No carbon dioxide was used in the production of this model.

  7. Bob Tisdale says:

    tallbloke says: “No carbon dioxide was used in the production of this model.”

    Thanks. I got a chuckle out of that this morning.

  8. tallbloke says:

    I’ve nailed it with HADSST3

    Bob, please could you email me a text file with hadsst3 in a single column, I’m having problems getting the data from KNMI into my spreadsheet. Thanks.

  9. tallbloke says:

    Huge thanks to Bob, I now have a nice looking plot with a correlation coefficient.

    I’ll do a write up my blog soon. So I’ll stop hogging bandwidth on Bob’s now.

    Thanks again mate.

  10. DB says:

    Speaking of centennial oscillations, Bohm has a paper out that looks at climate variability in central Europe over the last 250 years. The paper draws three conclusions:
    – there has been no increase of variability during the past 250 years in the region.
    – the last 30 years (dominated by GHG forcing) has no increase in variability.
    – there is a clear centennial oscillating structure for all three climate variables looked at.

    Changes of regional climate variability in central Europe during the past 250 years

  11. Pingback: On Dana1981’s Meaningless ENSO Exercise at SkepticalScience | Bob Tisdale - Climate Observations

  12. Pingback: On Dana1981’s Meaningless ENSO Exercise at SkepticalScience | Watts Up With That?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s